The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for building components that ingest external data, creating a potential surface for indirect prompt injection if the resulting plugins are integrated into automated agent workflows. 1. Ingestion points: External data enters via admin configuration models (references/admin-extension.md) and service webhooks (references/dependency-injection.md). 2. Boundary markers: The documentation encourages the use of AutoValidateAntiforgeryToken and model validation, although it does not explicitly define delimiters for separating untrusted data from agent instructions. 3. Capability inventory: The plugin framework grants access to the database via IRepository and INopDataProvider (references/data-access-and-ef.md), as well as network access via IHttpClientFactory (references/dependency-injection.md). 4. Sanitization: The skill advocates for robust sanitization practices, including the use of FluentValidation for input and IPermissionService for enforcing authorization boundaries.

nopcommerce